Health Education Research Advance Access originally published online on July 18, 2006
Health Education Research 2007 22(1):128-138; doi:10.1093/her/cyl056
Bystanding an abused child: testing the applicability of the stages of change construct
1 School of Psychology, Open University of the Netherlands, PO Box 2960, 6401 DL Heerlen, The Netherlands
2 Department of Health Promotion, University of Maastricht, PO Box 616, 6300 MD Maastricht, The Netherlands
* Correspondence to: A. N. Mudde. E-mail: aart.mudde{at}ou.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
A minority of child abuse cases is recognized by professionals, making the role of non-professional bystanders essential. The stages of change construct, as proposed by the Transtheoretical Model, may provide a useful approach to explain non-professional helping behavior. The objective of this study was to test the applicability of the stages of change construct by (i) assessing whether cognitive determinants distinguish between the stages, (ii) testing the predictive value of the stages for future helping behavior and (iii) examining the mediating role of the stages in the relation between previous and future helping behavior. Data of 126 adult non-professional bystanders were analyzed. Respondents were questioned by telephone or via an Internet questionnaire, at baseline and after a follow-up of about two months later. Attitude toward helping was significantly less positive in pre-contemplation than in the other stages, and self-efficacy expectations were significantly higher in preparation compared with the other stages. Moreover, baseline preparators were more likely to conduct future helping behavior than those in the two earlier stages. Finally, the stages of change mediated the relation between previous and future helping behavior. Initial support was found for the applicability of the stages of change construct for helping behavior by non-professional bystanders.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Child physical and emotional abuse and neglect, as well as child sexual abuse are generally regarded as serious problems. In most cases, child abuse is not a single incident [1], and the abuse tends to be take place over a prolonged period of time, often persisting over many years [e.g. 2–4]. In the United States, reports of child abuse to the public authorities have more than doubled, from 1.1 in 1980 to 2.3 million cases in 1993 [5]. In Canada, an increase of 125% was noted, from 9.64 reported cases per 1000 children in 1998 to 21.71 cases per 1000 children in 2003 [1]. There is a general consensus that only ‘the tip of the iceberg’ is being reported to child protection agencies, and that the magnitude of the problem may be much larger than is currently observed in clinical settings. This is illustrated by another Canadian study [6], which found that only 10% of the respondents who reported to be victims of any type of abuse had ever been in contact with child protection services. Moreover, Finkelhor [7] reviewed prevalence rates of child sexual abuse in non-clinical settings in 21 countries, and found figures ranging from 7% to 36% for women and 3% to 29% for men. These figures underline the potential of non-professionals to recognize child abuse. However, Trocmé et al. [1] concluded that 75% of the substantiated reports, constituting 40% of all reports, were made by professionals, indicating that reports from non-professionals may not be as valid as those from professional sources.
Many countries have introduced mandatory reporting systems for suspected child abuse by professionals, such as teachers and general practitioners, and numerous studies have tried to find out why professionals in these countries did or did not report abuse allegations [e.g. 8–10]. However, the evidence from this line of research probably does not contribute to an understanding of helping behavior by non-professional individuals, because the nature and determinants of professional and non-professional helping behavior may differ, as well as the contexts in which helping behavior is performed. Theory-driven studies addressing non-professional bystanders' willingness, ability and decisions to help in cases of suspected child abuse appear to be lacking.
The social-cognitive theoretical frame would seem to provide a useful approach to explaining bystanders' helping behavior, because the motives for behavior change are supposed to involve personal perceptions, such as suspicions of abuse, rather than actual emergencies, as described, for instance, in the ‘Model of the Intervention Process’ [11]. An increasing number of studies have provided empirical support for the applicability of this social-cognitive approach to explain health behaviors [for reviews, see 12, 13]. The theoretical background for the present study is a model originating from this approach: the attitude—social influence—self-efficacy (ASE) model [14]. This model integrates the principles of two theories: the Transtheoretical Model [15] and the Theory of Reasoned Action [16]. The Transtheoretical Model proposes the stages of change construct, which in the context of abuse implies that bystanders do not simply change their helping behavior overnight, but go through different stages to achieve sustained behavioral change. These stages are based on the time that an individual expects to elapse between the present moment and the intended behavior change, and are classified as pre-contemplation (not intending to help within the next 6 months), contemplation (intending to help within 6 months) and preparation (intending to help within 1 month). After the behavioral change has taken place (action), the changed behavior can be continued (maintenance) or discontinued (relapse) [17].
The stages of change are incorporated in the ASE model [14], together with the principal idea of the Theory of Reasoned Action [16], which postulates that the intention to change is directed by cognitive factors. The ASE model has evolved in several directions since Bandura's concept of self-efficacy expectations was added to it [for details, see 18]. The cognitive concepts of attitude, perceived social influence and self-efficacy expectations are acknowledged to direct movement through the stages of change. The cognitions are in turn determined by external variables, such as personality and biological factors, as well as by previous behaviors, experience and observations of the behavior of others. This model has been used as a theoretical basis for a number of studies of behavioral change [19–21]. Examining the differences between the stages, De Vries and Backbier [22] initially described a Ø-shaped pattern (see Fig. 1) in a population of Dutch adult smokers, where pre-contemplators perceived fewer advantages of behavior change than those in the other stages, and pre-contemplators and contemplators had lower self-efficacy expectations than those in the preparation or action stages. No significant differences were observed between the stages with respect to perceived social influence. In their study [22] and in later cross-sectional studies, these findings were confirmed for a range of behaviors [23–28]. Furthermore, a few longitudinal studies found differences in cognitions within stages to be predictive of stage transitions [14, 29, 30].
|
Because evidence for the applicability of the stages of change construct to helping behavior by non-professionals is lacking, the present study aimed to test the applicability of this construct for this specific behavior by conducting two of the four validation tests for stage models suggested by Weinstein et al. [31]. The first test comprises a cross-sectional comparison of characteristics of people in the different stages of change. In line with the Ø-shaped pattern depicted in Fig. 1, we hypothesized that beliefs would distinguish between the pre-contemplation stage on the one hand and the contemplation and preparation stages on the other, while self-efficacy would distinguish between the pre-contemplation and contemplation stages on the one hand and the preparation stage on the other. Moreover, based on earlier studies [22–29], we hypothesized that perceptions of social influence (norms and support) would not differ significantly between the stages. The second validation test involves the predictive value of the stages. Stage assignment at one point in time should be predictive of behavior assessed at a later moment [31]. In addition, we tested the underlying assumption of the social-cognitive theoretical frame, that the relation between previous behavior and future behavior is mediated by the stages of change.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Design
Each year on 10 October, under the auspices of the World Federation for Mental Health, special attention is paid to a particular worldwide mental health problem, to increase awareness, fight prejudices and inform populations. In 2002, the mental health problem was the effects of trauma and violence against children. Awareness-raising campaigns and other interventions were launched, adapted in each nation to the national or regional conditions (e.g. war, instability, etc.). In the Netherlands, a short-term media campaign called ‘child in trouble’ informed non-professional potential bystanders about the seriousness of child abuse and neglect. Interested people were directed to an Internet site and a telephone hotline where they could obtain more information and advice. The campaign was evaluated by studying exposure and changes in helping behavior among those who visited the website or called the hotline, using a one-sample pre–post measurement design. In the present study, the data of this study were analyzed to examine the applicability of the stages of change construct.
Recruitment
Two recruitment strategies were applied. First, all persons who called the hotline in the period between 10 and 16 October 2002 were asked to participate in a telephone interview if they knew a child in trouble. Second, visitors of the campaign website were invited to fill out an online electronic questionnaire if they knew a child in trouble. Website visitors who preferred to receive the electronic questionnaire by mail, could leave their e-mail address. Respondents were interviewed by professional interviewers who had received supplementary training. The follow-up assessment took place though the same channel as that used at baseline, 
2 months later (mean = 53 days, SD = 8.17).
Baseline questionnaire
The telephone and electronic questionnaires were identical, except for slight adaptations to the medium used. The baseline questionnaire assessed demographics of the respondent, characteristics of the possibly maltreated child and earlier helping behavior, exposure to the campaign, stages of change and psychosocial determinants.
Demographics
Questions were asked concerning the respondents' age, gender, ethnic background and education and how many children they had themselves.
Characteristics of the possibly maltreated child and previous helping behavior
The first question asked whether the respondent knew a maltreated child or suspected maltreatment. Age and gender of the child and the socioeconomic status and ethnic background of the child's family were then assessed, followed by the respondent's relation with the child (family relation, friend of own child, child of friend, child living in the neighborhood or knowing the child professionally), and how long the respondent had known the child. Finally, respondents were asked to list earlier attempts to help the child.
Exposure to the campaign
Respondents were asked how they knew of the hotline or the website and what type of information they had been seeking. They were also asked whether they had the campaign brochure in their possession and whether they had read it.
Stage of change
Stage of change was assessed by a series of questions. The first question was ‘Do you intend to help the child?’ If respondents answered ‘yes’ to this question, they were asked ‘Do you intend to help the child within 6 months?’ Respondents who answered ‘no’ or ‘I don't know’ to Question 1 or 2 were labeled as pre-contemplators. If respondents answered yes to Question 2, they were asked ‘Do you intend to help the child suspected of being abused within 30 days?’ Respondents answering this question in the negative were labeled contemplators. If they answered yes, they were labeled preparators.
Psychosocial determinants
The following psychosocial determinants were assessed: attitude, social influence and self-efficacy. All items used statements and assessed the level of agreement with each statement by using five answering categories, ranging from ‘totally agree’ to ‘totally disagree’. Attitude was assessed by six statements: (i) ‘Raising a child is the parents' business’, (ii) ‘If there is a suspicion of abuse, family members or neighbors are allowed to interfere, even if parents did not ask them to’, (iii) ‘If you suspect that a child is in trouble, you have to help’, (iv) ‘I feel responsible for helping a child in trouble’, (v) ‘If I help an allegedly abused child, it will make me feel good about myself’ and (vi) ‘If I help an allegedly abused child, I am sure I am doing something good’. The items were summed to form an attitude scale (
= 0.74). Social influences were operationalized by means of two of the three concepts suggested by the ASE model: subjective norm and social support. These concepts were assessed by single items: subjective norm by ‘People who are important to me think I should help an allegedly abused child’ and social support by ‘I expect to receive support from my social environment when I help an allegedly abused child’. Self-efficacy was measured along the dimensions of perceived difficulty and perceived ability, by single items: perceived difficulty by ‘It would be difficult for me to help an allegedly abused child’ and perceived ability by ‘I am able to help an allegedly abused child’.
Follow-up questionnaire
Behavior
The follow-up questionnaire assessed the actions undertaken by respondents to help the child suspected of being abused. Examples of self-reported helping behavior are contacting professionals who work with children, helping out in the household in which the child lives and showing interest in drawings the child made.
Analyses
Demographic variables, characteristics of the possibly maltreated child, previous helping behavior and exposure to the campaign were tested as possible confounders in the relation between cognitive factors and stages of change and between stages and behavior in the follow-up.
Differences between dropouts and respondents were analyzed by means of logistic regression to determine possible selection. A forward stepwise model was constructed to control for overlapping variables.
To enable various scales to be compared in one figure, z-scores were computed for the cognitive variables.
The first validity test used one-way analysis of variance (ANOVA) with post hoc contrast with Sidak correction for multiple comparisons, with stages of change as the fixed factor and the cognitive factors as independent variables.
The second validity test was a logistic regression analysis which examined whether the transition toward the action stage could be predicted by the baseline stage distribution and whether this relation would persist when controlling for previous helping behavior.
Finally, we tested whether the stages of change mediated in the relation between previous and future behavior by examining whether the addition of the stages of change to the regression would eliminate the significant relation between previous and future behavior. If this was the case, a Sobel test statistic [32] was calculated to confirm the mediating effect.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Characteristics of the sample
Of the 1019 persons who called the telephone helpline, 168 people said that they knew a child in trouble and were willing to participate in the telephone interview. Of these people, 147 (88%) were interviewed. Reasons for non-response were that subjects could not be reached by telephone during the interview period (n = 15), gave a wrong number (n = 3) or had language problems (n = 3). The website was visited
9000 times (home page hits) and 197 people (2.2%) filled out the online version of the questionnaire. Of the total group of 344 respondents, a random sample of 51 (15%) served as a test-effect control group and only received the follow-up questionnaire. They were therefore not useful for the goal of the present study. For the purpose of this study, it was necessary to select a sample consisting of adult non-professional bystanders of a child suspected of being abused. Therefore, of the remaining 293 respondents, 165 (57%) were excluded for the following reasons: 119 respondents reported that they did not know a maltreated child or did not suspect maltreatment, 39 respondents said that they were involved professionally, four respondents were aged 18 years or younger and three respondents indicated that they had themselves been subjected to abuse as children. Two questionnaires had >10% missing values and were therefore not fit for analysis. The remaining 126 subjects had a mean age of 40.94 years (SD = 12.45); 80.2% were female, and 96.8% were of native Dutch origin. The majority (40.7%) reported secondary education granting admission to higher education level as their highest completed education, while 30.1% reported higher level training (higher vocational school or university), 21.9% mentioned a lower level secondary school and 7.3% primary school or other. Finally, 76.2% reported that they had children of their own, with an average of 1.54 (SD = 0.88) boys and 1.48 (SD = 0.75) girls. At follow-up, data of 93 (73.8%) respondents were gathered. Dropout analysis showed that the only difference between dropouts and follow-up respondents concerned the way they had participated: 44% of those who filled out the electronic questionnaire dropped out versus 8% of the telephone interview group (OR = 0.11, P < 0.001). There were no differences in demographic variables or exposure measures.
Characteristics of the possibly maltreated child and previous helping behavior
The mean age of the children suspected of being abused was 9.44 years (SD = 4.65) and 52.4% were boys. In 38.9% of the cases, the child's family had a low socioeconomic status, and 15.1% of the children were of non-Dutch origin. The reported relationship between the respondents and the child was parent in 15.1% of the cases, uncle or aunt in 13.5%, grandparent in 12.7% and brother or sister in 3.2%. Non-family relationships included being a neighbor of the child (16.7%), the child being a friend of the respondent's own child (6.3%) or other (32.5%). Respondents had known the child for an average of 82.31 months (SD = 61.29). The most frequently reported type of abuse was neglect (38.9%), while physical abuse was reported by 17.5%. Sexual abuse was mentioned by 1.6% of the respondents. The combination of physical abuse and neglect was also often mentioned (34.9%), while combinations with sexual abuse were less frequent (sexual abuse and physical abuse 1.6%; sexual abuse, neglect and physical abuse 5.6%). With respect to previous helping behavior, >84% (N = 106) of the respondents reported to have acted earlier to help the child.
Distinguishing between stages
The test for possible confounders did not reveal consistent confounders, so there was no need to include covariates in further analyses. Of the 126 respondents, 35 (27.7%) were in the pre-contemplation stage, while 16 (12.7%) were in the contemplation stage and 75 (59.5%) in the preparation stage. The results of the first validity test on the differences between the various baseline stages in terms of cognitive factors are summarized in Table I. The expected attitudinal difference between pre-contemplators on the one hand and contemplators and preparators on the other was partly confirmed: pre-contemplators had lower scores than contemplators and preparators, and the difference between contemplators and preparators was negligible. However, only the difference between pre-contemplators and preparators was significant. With respect to self-efficacy, it was expected that preparators would have higher scores than pre-contemplators and contemplators. This expectation was partly supported for the ability dimension of self-efficacy, which differed significantly between preparators and contemplators, and did not differ between pre-contemplators and contemplators. However, the difference between preparators and pre-contemplators was not significant. No differences between the stages were found for perceived difficulty. The differences between the stages with respect to social influences were consistent with the Ø pattern, and not significant. Fig. 2 visualizes the differences between the stages.
|
|
Prediction of future behavior
Of the baseline pre-contemplators (N = 24), 25% proceeded to the action stage, while 60% of the contemplators (N = 16) did so. Preparators (N = 59) were most likely to act: 90% of them reported helping behavior at follow-up. The baseline stage division between preparators and the two earlier stages proved to be a powerful predictor of helping behavior at follow-up, even when controlled for baseline helping behavior. Baseline preparators were significantly more likely to proceed to action than pre-contemplators (OR = 11.03, P < 0.001). Contemplators did not differ significantly from pre-contemplators in terms of the likelihood of proceeding to actual helping behavior (OR = 2.91, ns) (Nagelkerke R2 = 0.29).
Previous helping behavior was significantly related to future behavior (OR = 2.93, P < 0.05), but this relation disappeared when stage was entered, indicating the supposed mediating role of the stages of change between previous helping behavior and future helping behavior. This mediating role was confirmed by a Sobel test (Sobel test statistic = 2.59, P < 0.01).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
The present study examined the applicability of the stages of change construct to bystanders' helping behavior in cases of suspected child abuse. A literature search showed that this study is likely to be the first to report on the relation between psychosocial determinants and stages of change in this field. Our analysis included data of 126 participants of a Dutch evaluation study, and two tests to examine the validity of the stages of change model, as suggested by Weinstein et al. [31]. The study also assessed the mediating role of the stages of change in the relation between previous and future helping behavior.
The results of the first validity test, implying that people in the various stages of change can be distinguished by cognitive factors, statistically confirmed several hypotheses derived form the Ø-shaped pattern (Fig. 1). The expected difference in attitude between pre-contemplators on the one hand and contemplators and preparators on the other hand was confirmed for preparators, who scored significantly higher than pre-contemplators, but not for contemplators, although their scores were higher. A statistical explanation might lie in the relatively small number of contemplators in our sample. A conceptual explanation might involve the time frame of contemplation, which may be unreasonable for behaviors requiring urgency or taking place during specific developmental stages (e.g. young age or pregnancy) [33, 34]. In these cases, respondents might be more inclined either to report refraining from action (pre-contemplation) or to act in the short term (preparation), than planning a change in 6 months.
The hypothesized higher self-efficacy score of preparators compared with contemplators was confirmed, while no difference was found with pre-contemplators. The expected pattern for the self-efficacy dimension of perceived difficulty was not found. A methodological explanation might be the instrument's one-item format, although it has been shown that a single-item instrument for perceived ability toward smoking cessation performed as well as a multi-item scale [35]. A theoretical explanation might be that the two self-efficacy constructs delineate different dimensions, which may not or may co-occur: respondents may experience themselves as being able to help and at the same time realize how difficult helping behavior will be. The smoking cessation study referred to above [35] concluded that perceived ability was a better operationalization of self-efficacy than perceived difficulty in terms of predicting smoking cessation. Additional research into the best operationalization of self-efficacy in the field under study here is needed to confirm these explanations.
The hypothesis, derived from other studies [22–29], that perceptions of social influence, including social norms and social support, would not differ between the three stages was supported by the findings of our study. Although the perceptions of social influence were stronger in each of the successive stages, the differences between the stages were not significant. In conclusion, our cross-sectional findings suggest that our study passed the first validity test suggested by Weinstein et al. [31].
As regards the second validity test, the findings support the predictive value of the stages of change construct for behavior later in time. The transition to the action stage was most likely for preparators and least likely for pre-contemplators. The baseline distinction between preparation and the other two stages was a powerful predictor of behavior change, even when controlling for previous behavior. The limited size of the group of contemplators may be the reason why we did not find a significant difference between pre-contemplators' and contemplators' likelihood to proceed to helping behavior. The test of the underlying social-cognitive theoretical framework, implying that the influence of previous behavior on future behavior is mediated by the stage of change, confirmed this mediating role, indicating that the theoretical frame is applicable to bystanders' helping behavior.
Several limitations of the present study must be addressed. As noted before in relation to the self-efficacy construct, the novel field of research means that there is a lack of validated instruments. Both qualitative and quantitative validation studies are needed to identify salient attitude and self-efficacy beliefs and to assess the psychometric properties of scales [16]. In addition, the high rate of reported previous helping behavior indicates that many respondents may have been engaged in helping behavior at baseline. Because present helping behavior and its duration were not assessed at baseline, subjects in the action or maintenance stage could not be identified. However, any possible effect of this omission was probably limited, since the correction for previous behavior, which partly suppresses the influence of baseline behavior, did not affect the relation between stages and future helping behavior.
Apart from measurement imperfections, the internal validity of the present study may also have suffered from social desirability bias. However, existing evidence that data collected by Internet surveys have higher predictive validity than data collected by means of telephone samples [36], and the higher dropout rates among Internet respondents in the present study, suggest that internet respondents experience less pressure to participate and therefore may answer more honestly and less according to what they think is expected of them.
The external validity of the study may have been affected by selection bias. The sample was selected from a group of people who responded to a campaign with information seeking behavior, who knew a child in trouble and who were willing to participate in the present study. This sample is likely to be more personally involved and motivated to act than the general population. Moreover, the two sampling methods probably selected different samples from the total population of non-professional bystanders. For example, a study among 45 000 Dutch employees revealed that men, more highly educated employees and younger employees were more likely to respond through the Internet than others [37]. As a consequence, the present conclusions may not be applicable to all non-professional bystanders, because those bystanders who did not seek information or did not want to participate were not represented. However, the effort made in this study to recruit respondents via different channels may have reduced medium-dependent self-selection bias and thus increased the representativeness of the sample and the external validity of the study. Finally, generalizing the findings to different countries may be hampered by differences in public opinion toward child abuse, legislation and reporting systems. The Netherlands, like many European countries, has not introduced a mandatory reporting system, and legislation on child abuse is less strict than for instance that in the United States [9]. This may have resulted in differences in public opinion and individual attitudes.
The present findings may have several implications for practice and research. Even a low-intensity intervention of short duration was found to be able to target pre-contemplators and to stimulate them to seek information. The findings showed that the attitude of this group before the intervention was relatively negative. This supports the widely held opinion that the societal response to child abuse might benefit from awareness-raising interventions addressing attitudes [8]. However, the potential of this type of intervention is likely to be limited to the promotion of stage transitions from pre-contemplation to contemplation only, and not to the more action-inclined stages. Transitions to these later stages are more likely to be achieved by self-efficacy enhancing targets. The conclusion emerges that not all respondents will profit from information either in terms of changing their attitudes or in terms of changing their self-efficacy, but that information should ideally be stage specific [19, 29]. Media campaigns are probably more effective in evoking transitions between the earlier stages, since their agenda-setting potential has been shown, while transitions to the action-inclined stages are most likely to profit from combinations with other interventions, such as skills training and legislation. Lamond [38] underpinned this view by showing that legislation compelling teachers to report suspected abuse led to an increase in the number of cases reported by teachers, while the accompanying mass media campaign aimed at the general public did not result in behavior change.
Although the present study provides preliminary support for the applicability of the stages of change construct to non-professional bystanders' helping behavior toward abused children, Weinstein et al. [31] suggest two additional critical tests. First, prospective studies are needed to test the assumption that different stages involve different predictors of stage transitions [17, 31]. Since the present study provided no follow-up data on the stages of change, it was not possible to apply this test. Second, the ultimate tests of stage theories involve experiments with matched and mismatched interventions [31]. However, in view of the ethical problems of withholding relevant information from bystanders of abused children, experiments probably have to be limited to vignette studies involving hypothetical cases.
Given the point prevalence of child abuse, an important number of children are immediately in need of help or care. The UN Convention of the Rights of the Child [39], ratified by almost all nations in the world, states that societies need to mobilize their potential to provide this help and care. The present study has shown that non-professional bystanders, as part of this potential, can be differentiated along meaningful theoretical dimensions, i.e. stages of change and psychosocial determinants of behavior, providing cues for interventions. Practitioners in the field may apply these findings in their work to activate non-professionals and to stimulate helping behavior by considering the stages their target group is in and then tailoring interventions to their respective needs. Since more than one-fourth of our sample of active information seekers was in the pre-contemplation stage, it can be assumed that this percentage is much higher in the general population who did not respond to the campaign. Therefore, the first step in stimulating bystanders of child abuse to proceed through the stages of helping behavior should be to target the attitudes of pre-contemplators by means of mass media interventions. This type of intervention should then be followed up by more intensive and tailored interventions aimed at increasing skills and policy measures to enforce behavior change.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
None declared.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
This study was partially funded by grants from the Netherlands Organization for Health Research and Development (ZonMw) and the National Foundation for Mental Health (NFGV). We are grateful to Marjolein Nieuwenkuiper and Peter van Cooten of the telephone help service for psychological and social problems (Korrelatie), Leon Kolenburg (Department of Health Promotion, University of Maastricht), Adrianne Dercksen, Jurjun Procee and Ingvil van de Lisdonk Dutch National Fund for Public Mental Health (NFGV), and the interview facilities of Center for Data and Information Management (MEMIC) Centre for Data and Information Management, for their contribution to data gathering and processing.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
1. Trocmé N, Fallon B, MacLaurin B, et al. Canadian Incidence Study of Reported Child Abuse and Neglect—2003: Major Findings.Ottowa, Canada: Minister of Public Works and Government Services 2005.
2. Finkelhor D, Hotaling G, Lewis IA, et al. Sexual abuse in a national survey of adult men and women: prevalence, characteristics, and risk factors. Child Abuse Negl 1990 14:19–28.[CrossRef][Web of Science][Medline]
3. Draijer N. Seksuele traumatisering in de jeugd -Gevolgen op lange termijn van seksueel misbruik van meisjes door verwanten. [Sexual Traumatization in Childhood—Long-Term Consequences of Sexual Abuse of Girls by Relatives.].Amsterdam, The Netherlands: SUA 1990.
4. Russell DEH. The prevalence, trauma, and sociocultural causes of incestuous abuse of females: a human rights issue. In Kleber RJ, Figley CR, Gersons BP (Eds.). Beyond Trauma, Cultural and Societal Dynamics.New York: Plenum Press 1995 pp. 171–86.
5. US Department Health Human Services. Child Maltreatment 1993: Reports from States to the National Center on Child Abuse and Neglect.Washington, DC: Government Printing Office 1995.
6. MacMillan HL, Jamieson E, Walsh C. Reported contact with child protective services among those reporting child physical and sexual abuse: results from a community survey. Child Abuse Negl 2003 27:1397–408.[CrossRef][Web of Science][Medline]
7. Finkelhor D. The international epidemiology of child sexual abuse. Child Abuse Negl 1994 18:409–17.[CrossRef][Web of Science][Medline]
8. Bross JD, Miyoshi TJ, Miyoshi PK, et al. World Perspectives on Child Abuse—the Fourth International Resource Book.Denver, CO: Kempe Children's Center, University of Colorado School of Medicine 2000.
9. Gilbert N. Conclusion: a comparative perspective. In Gilbert N (Ed.). Combating Child Abuse—International Perspectives and Trends.New York: Oxford University Press 1997 pp. 232–40.
10. Unen AV. Meldingssystemen inzake kindermishandeling in Noord- en West-Europaen de Verenigde Staten van Amerika. [Reporting Systems for Child Abuse in Northern and Western Europe and the United States of America.].Utrecht, The Netherlands: NIZW 1997.
11. Latané B and Darley JM. The Unresponsive Bystander: Why Doesn't He help?New York: Meredith Corporation 1970.
12. Godin G and Kok GJ. The theory of planned behavior: a review of its applications to health-related problems. Am J Health Promot 1997 11:87–98.
13. Armitage CJ and Conner M. Efficacy of the theory of planned behavior: a meta-analytic review. Br J Soc Psychol 2001 40:471–99.[CrossRef][Web of Science][Medline]
14. De Vries H and Mudde AN. Predicting stage transitions for smoking cessation applying the Attitude—Social Influence—Efficacy Model. Psychol Health 1998 13:369–85.[Web of Science]
15. Prochaska JO and Diclemente CC. Stages and processes of self-change of smoking: toward an integrative model of change. J Consult Clin Psychol 1983 51:390–5.[CrossRef][Web of Science][Medline]
16. Fishbein M and Ajzen I. Belief, Attitude, Intention and Behavior.New York: Willey 1975.
17. Prochaska JO, DiClemente CC, Norcross JC. In search of how people change: applications to addictive behaviors. In Marlatt G and Van den Bos GR (Eds.). Addicitive Behaviors: Readings on Etiology, Prevention, and Treatment.Washington, DC: American Psychological Association 1997 pp. 671–96.
18. Kok GJ, De Vries H, Mudde AN, et al. Planned health education and the role of self-efficacy: Dutch research. Health Educ Res 1991 6:231–8.
19. Ausems M, Mesters I, De Vries H, et al. Short-term effects of a randomized computer-based out-of-school smoking prevention trial aimed at elementary schoolchildren. Prev Med 2002 34:581–9.[CrossRef][Web of Science][Medline]
20. Bolman C and De Vries H. Psycho-social determinants and motivational phases in smoking behavior of cardiac inpatients. Prev Med 1998 27:738–47.[CrossRef][Web of Science][Medline]
21. De Nooijer J, Lechner L, De Vries H. Social psychological correlates of paying attention to cancer symptoms and seeking medical help. Soc Sci Med 2003 56:915–20.[CrossRef][Web of Science][Medline]
22. De Vries H and Backbier MPH. Self-efficacy as an important determinant of quitting among smoking pregnant women: the Ø-phenomenon. Prev Med 1994 23:167–74.[CrossRef][Web of Science][Medline]
23. De Vries H, Mudde AN, Dijkstra A, et al. Differential beliefs, perceived social influences and self-efficacy expectations among smokers in various motivational phases. Prev Med 1998 27:681–9.[CrossRef][Web of Science][Medline]
24. Brug J and Van Assema P. Factors differentiating between stage of change for fat reduction. Appetite 1995 24:296.
25. Lechner L and De Vries H. Participation in an employee fitness program: determinants of high adherence, low adherence and drop-out. J Occup Environ Med 1995 37:429–36.[Web of Science][Medline]
26. Lechner L and De Vries H. Starting participation in an employee fitness program: attitudes, self-efficacy and social influence. Prev Med 1995 24:627–33.[CrossRef][Web of Science][Medline]
27. Oostveen T, Knibbe RA, De Vries H. Social influences on young adults' alcohol consumption; norms, modeling, pressure, socializing and conformity. Addict Behav 1996 21:187–97.[CrossRef][Web of Science][Medline]
28. Wortel E, De Vries H, De Geus GH. Lessons learned from a community campaign on child safety in The Netherlands. Fam Community Health 1995 18:60–77.
29. De Vries H, Mudde AN, Dijkstra A. The Attitude—Social Influence—Efficacy Model: an application to the prediction of motivational transitions in the process of smoking cessation. In Norman P, Abraham C, Conner M (Eds.). Understanding and Changing Health Behaviour: from Health Beliefs to Self-Regulation.Amsterdam, The Netherlands: Harwood Academic 2000 pp. 165–87.
30. Velicer WF, Norman GJ, Fava IL, et al. Testing 40 predictions from the Transtheoretical Model. Addict Behav 1999 24:455–69.[CrossRef][Web of Science][Medline]
31. Weinstein ND, Rothman AJ, Sutton SR. Stage theories of health behavior: conceptual and methodological issues. Health Psychol 1998 17:290–9.[CrossRef][Web of Science][Medline]
32. Preacher KJ and Leonardelli GJ. Calculation for the Sobel test: an interactive calculation tool for mediation tests [computer software]. Available at: http://www.unc.edu/preacher/sobel/sobel.htm, 2001. Accessed: 25 August 2005.
33. Bandura A. The anatomy of stages of change. Am J Health Promot 1997 12:8–10.[Web of Science][Medline]
34. Sutton SR. Back to the drawing board? A review of applications of the transtheoretical model to substance use. Addiction 2001 96:175–86.[CrossRef][Web of Science][Medline]
35. Mudde AN, Kok GJ, Strecher VJ. Self-efficacy as a predictor for the cessation of smoking: methodological issues and implications for smoking cessation programs. Psychol Health 1995 10:353–67.
36. Chang L. A comparison of samples and response quality obtained from RDD telephone survey methodology and Internet survey methodology. Diss Abstr Int Sect B Sci Eng 2002 62:3845.
37. Taris T. Dataverzameling via internet: een update. [Web-Based Data Collection: an Update.] Gedrag Organ 2005 18:181–95.
38. Lamond DAP. The impact of mandatory reporting legislation on reporting behavior. Child Abuse Negl 1989 13:471–80.[CrossRef][Web of Science][Medline]
39. Unicef. Convention of the Rights of the Child. Available at: http://www.unicef.org/crc/crc.htm. Accessed: 20 August 2005.
Received on September 9, 2005; accepted on June 2, 2006
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  V. L. Banyard, R. P. Eckstein, and M. M. Moynihan Sexual Violence Prevention: The Role of Stages of Change J Interpers Violence, January 1, 2010; 25(1): 111 - 135. [Abstract] [PDF]
|
|
|
|
 |
|
 S. J. Potter, M. M. Moynihan, J. G. Stapleton, and V. L. Banyard Empowering Bystanders to Prevent Campus Violence Against Women: A Preliminary Evaluation of a Poster Campaign Violence Against Women, January 1, 2009; 15(1): 106 - 121. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||